JP5565357B2 - Equipment diagnostic device, equipment diagnostic method, equipment diagnostic program, and computer-readable recording medium recording the same - Google Patents

Description

  The present invention relates to an equipment diagnosis apparatus, an equipment diagnosis method, an equipment diagnosis program, and a computer-readable recording medium on which the equipment diagnosis program is used to diagnose equipment of a steel plant or other production system.

  Conventionally, as a method of monitoring an abnormality of equipment that is sequence-controlled by a programmable logic controller (hereinafter referred to as “PLC”), a method of comparing “actual (normal) operating time” with “actual operating time” Is often used. Specifically, there are many cases in which a time from a start signal (for example, activation signal on) to an end signal (for example, position detection sensor on) of a certain operation is measured, and an abnormality is determined when it is delayed by a certain threshold or more.

  For example, in Patent Document 1, in an automated facility that sequentially performs a plurality of steps, when compared with a predetermined operation time at a normal time from the operation start to the operation end of each step, when it is determined to be abnormal , It is described to display to the operator which single process in the series of processes is abnormal. Patent Document 2 describes that an external device automatically extracts and compares an on / off pattern of a signal serving as a determination reference from an existing sequence control device. All of them are assumed to be equipment that performs basically uniform operation and has almost no variation in operation time.

  In addition, Patent Document 3 describes that, in order to diagnose an abnormality in equipment even when there is a variation in operating time, a statistical amount of operating time is obtained and the reference value and the actual value are compared with each other. ing. However, in this diagnostic device, it is assumed that the operation time takes a substantially constant value, and although a variation can be allowed to some extent, a signal that does not repeat ON / OFF at a predetermined timing and the appearance order are different. In such a case, it is specified that it is not a target for feature extraction.

  Patent Document 4 collects equipment state data indicating the state of each equipment and product state data indicating the state of the produced product, and classifies the collected data into cases where the product is normal and abnormal. Then, for each of the classified normal state data and abnormal state data, the feature amount is detected by using a predetermined operation of each facility as a trigger, and the feature amount in the normal state and the feature amount in the abnormal state are detected. It is described that each feature amount is registered in the database only for those that have a difference of a predetermined threshold or more by comparison, and the abnormality diagnosis is performed by comparing the feature amount extracted from each facility during operation with the database. Yes.

Japanese Patent Laid-Open No. 7-271431 Japanese Patent Laid-Open No. 5-181511 JP 2008-226006 A JP 2010-277199 A

  In a facility that performs uniform operation under constant conditions, the conventional technology as described in Patent Documents 1 to 3 can be applied. However, the actuator drive source condition (hydraulic pressure) depends on the operation conditions as in a steel plant. If the property (material, weight, etc.) of the contacted object changes when it is operated while it is in contact with the object, the operating time is the same sequence operation. In some cases, large variations may occur, and the conventional technology cannot be applied.

  Further, in the technique described in Patent Document 4, when the quality of a product can be determined based on the equipment state data and the product state data, it is possible to detect a behavior change of a feature amount having a high degree of influence on the quality. Features with low degrees may be overlooked. However, since it is a necessary component of the production process, even if slight behavioral changes in the feature quantity do not affect the quality, there are many that are directly connected to the production stop as soon as it becomes inoperable. It is also necessary to detect a feature amount having a low influence degree.

  Therefore, in the present invention, a facility diagnosis device, a facility diagnosis method, a facility diagnosis program, and an apparatus diagnosis program that eliminates variations in feature amounts such as operation time due to a difference in operating conditions with respect to a facility to be diagnosed, and enables accurate diagnosis. An object is to provide a recorded computer-readable recording medium.

  The facility diagnosis apparatus of the present invention includes a diagnosis item determination means for determining a facility diagnosis item from a trigger signal based on information on a predefined facility diagnosis item and a trigger signal of each facility diagnosis item, and a determined facility diagnosis item Are divided into groups based on group classifications defined in advance by operating conditions, feature quantity extraction means for extracting feature quantities according to equipment diagnosis items for each group, and equipment diagnosis items for each group. And a diagnostic means for performing facility diagnosis by comparing the corresponding feature amount with a normal feature amount reference value.

  Further, the facility diagnosis method of the present invention is to determine the facility diagnosis item from the trigger signal based on the information on the trigger signal of the facility diagnosis item and each facility diagnosis item defined in advance, the determined facility diagnosis item, When the group is divided into groups based on the group classification defined in advance by operating conditions, the feature quantity corresponding to the equipment diagnosis item is extracted for each group, and the feature quantity corresponding to the equipment diagnosis item for each group is normal The facility diagnosis is performed by comparing with the feature amount reference value.

  According to these inventions, the equipment diagnosis items determined from the trigger signal are grouped based on the group classification defined in advance by the operation conditions, and the feature amount according to the equipment diagnosis items for each group is normal. By comparing with the feature value reference value, facility diagnosis can be performed while eliminating variations in feature values such as operation time due to differences in operation conditions.

  The facility diagnosis apparatus of the present invention includes a statistical processing unit that performs statistical processing according to a facility diagnosis item for each group based on the feature amount extracted by the feature amount extraction unit, and a feature extracted by the feature amount extraction unit. A diagnosis target database for registering the quantity and the statistics processed by the statistical processing means, and a reference value database for registering the normal feature quantity reference value and the statistic reference value according to the equipment diagnosis item for each group Further, it is desirable that the diagnosis means further performs facility diagnosis by comparing a statistic corresponding to the facility diagnosis item for each group with a statistic reference value in a normal case.

  As a result, not only the feature value but also the statistical value obtained by the statistical processing corresponding to the equipment diagnosis item for each group based on this feature value is compared with the standard value of the statistic value in the normal case. The equipment diagnosis can be performed by eliminating the variation of the feature amount such as the operation time.

  In addition, the facility diagnosis apparatus of the present invention calculates a statistic reference value using only those that have no difference greater than a predetermined value as a result of comparison of the statistic by the diagnostic means, and updates the reference value database. It is desirable to include a calculation means. Thereby, the statistic reference value in the normal case is automatically updated, and it becomes possible to perform more reliable equipment diagnosis.

  Further, the facility diagnosis program of the present invention includes a diagnosis item determination means for determining a facility diagnosis item from a trigger signal based on information related to a predefined facility diagnosis item and a trigger signal of each facility diagnosis item, and the determined facility Grouping means that divides diagnosis items into groups based on group classifications defined in advance by operation conditions, feature amount extraction means that extracts feature amounts according to facility diagnosis items for each group, and facility diagnosis for each group This is for causing the computer to function as a diagnostic means for performing facility diagnosis by comparing the characteristic amount corresponding to the item with the normal characteristic amount reference value. According to the computer which executed this program, the same operation and effect as the above-mentioned equipment diagnostic device of the present invention can be produced. Further, the facility diagnosis program can be provided by a computer-readable recording medium on which the facility diagnosis program is recorded, and can be read and executed by the computer.

(1) A facility diagnosis item is determined from a trigger signal based on information on a pre-defined facility diagnosis item and a trigger signal of each facility diagnosis item, and the determined facility diagnosis item is defined in advance according to operating conditions. By dividing into groups based on classification, extracting feature quantities according to equipment diagnosis items for each group, and comparing feature quantities according to equipment diagnosis items with normal feature quantity reference values for each group With the configuration for performing facility diagnosis, facility diagnosis can be performed by eliminating variations in feature amounts such as operation time due to differences in operating conditions, and accurate diagnosis can be performed.

(2) Perform statistical processing according to equipment diagnosis items for each group based on the extracted feature values, register the extracted feature values and processed statistics in the diagnosis target database, and install equipment diagnosis items for each group The feature value reference value and the statistic reference value in the normal case according to the standard are registered in the reference value database, and the statistic according to the equipment diagnosis item for each group is compared with the statistic reference value in the normal case. The equipment is diagnosed by the system, and not only the feature quantity but also the statistical quantity obtained by the statistical processing according to the equipment diagnosis item for each group eliminates the variation of the feature quantity such as the operation time due to the difference in operating conditions. Diagnosis can be performed, and more accurate diagnosis can be performed.

(3) As a result of comparison of statistics, the statistics reference value is calculated using only those that do not differ by more than a predetermined value, and the reference value database is updated. It becomes possible to perform more reliable equipment diagnosis.

It is explanatory drawing which shows the operation support system structure of the steel plant as a production system in embodiment of this invention. It is a block block diagram of the equipment diagnostic computer of FIG. It is a flowchart which shows the process of the equipment diagnostic computer of FIG.

  FIG. 1 is an explanatory diagram showing an operation support system configuration of a steel plant as a production system according to an embodiment of the present invention, and FIG. 2 is a block configuration diagram of an equipment diagnosis computer of FIG.

  In FIG. 1, a steel plant support system according to an embodiment of the present invention includes a control valve, a solenoid valve, a motor, various sensors that measure flow rate, pressure, temperature, and the like that constitute each process line of the steel plant. PLC (programmable logic controller) 2 as a control device (sequencer) that controls each facility 1 to be sequence-controlled, a general-purpose control network 3 to which the PLC 2 is connected, and a state of each facility 1 flowing through the general-purpose control network 3 It comprises a data collection computer 4 that collects equipment state data to be shown, an equipment diagnosis computer 5 as an equipment diagnosis device that diagnoses each equipment 1, and the like.

  The facility diagnosis computer 5 is a computer that executes a facility diagnosis program recorded on a computer-readable recording medium (not shown). As shown in FIG. 2, a master information registration unit 10, a data collection unit 11, a diagnosis item determination unit. 12, grouping means 13, feature quantity extraction means 14, statistical processing means 15, diagnosis means 16, statistical quantity reference value calculation means 17, output means 18 and the like. The facility diagnosis computer 5 includes a master information database (DB) 20, a performance database (DB) 21, a diagnosis target database (DB) 22, a reference value database (DB) 23, and the like in a storage device (not shown). Has been built.

  In the master information DB 20, the definition of the equipment 1 to be diagnosed, the definition of equipment diagnosis items (for example, monitoring the operating time of the A cylinder and B furnace temperature) indicating the pre-defined diagnostic processing contents, and each equipment diagnosis item Of the trigger signal (for example, A cylinder ascending command, A cylinder ascending limit, B furnace temperature rising command, B furnace temperature results, etc.) Definition, definition of statistical processing method for each equipment diagnosis item, definition of group classification defined in advance by operation conditions (for example, steel grade (hardness, etc.) and size (sheet width, thickness, weight, etc.)) Master information is registered. The master information registration means 10 registers these master information in the master information DB 20.

  The data collection unit 11 collects current facility state data collected by the data collection computer 4 or the like. What information is collected is determined based on the definition of the facility 1 to be diagnosed registered in advance in the master information DB 20. The equipment state data is various process data results (for example, operation conditions, trigger signals, etc.) that can be grasped in real time such as various results and sequences during operation of each equipment 1. These pieces of equipment state data collected by the data collecting unit 11 are registered in the result DB 21. The equipment state data is not limited to being collected by the data collection computer 4, but may be configured to be collected by another method.

  The diagnosis item determination means 12 determines an equipment diagnosis item from the trigger signal based on information on the equipment diagnosis item registered in the master information DB 20 and the trigger signal of each equipment diagnosis item. The diagnosis item determination means 12 always checks the state of each trigger signal in various process data results, and determines the facility diagnosis item corresponding to this trigger signal when the trigger signal is activated. For example, when the A cylinder ascending command works as a trigger signal, the diagnosis item determination means 12 determines that the operation time monitoring of the A cylinder is performed as an equipment diagnosis item corresponding to the trigger signal. The diagnostic item determination means 12 can be configured to check not only the preset trigger signal but also the state of the trigger signal by a determination flow.

  The grouping means 13 divides the equipment diagnosis items determined by the diagnosis item determination means 12 into groups based on the group classification defined by the operation conditions registered in the master information DB 20. Tables 1 and 2 show examples of group divisions defined by operating conditions. Table 1 shows an example of group classification defined by the coil width and coil weight, which are operation conditions, for monitoring MD support attachment / detachment time as an equipment diagnosis item. Table 2 shows an example of the group classification defined by the steel type and the coil plate thickness as the operation conditions for the PR reduction time as the equipment diagnosis item. In the example of Table 1 and Table 2, although it is a two-dimensional table, it may be grouped in one dimension, three dimensions or more according to the number of influencing factors. Moreover, it is also possible to adopt a configuration in which grouping is performed using a discrimination flow instead of grouping by a table.

  The feature quantity extraction unit 14 extracts feature quantities corresponding to the equipment diagnosis items for each group. The extracted feature amount is registered in the diagnosis target DB 22. The feature amount is a result obtained by executing the facility diagnosis item, and the feature amount is extracted based on the definition of the feature amount extraction method for each facility diagnosis item registered in the master information DB 20. For example, if the equipment diagnosis item is monitoring the operating time of the A cylinder, the feature amount is the time for one operation (the upper limit switch (LS) is turned on from the time when the A cylinder ascent command is turned on (the A cylinder ascent limit) Time to reach).

  As another example of the feature quantity, when the equipment diagnosis item is pattern control monitoring (such as those that are directly controlled) such as temperature, speed, pressure, etc., the target value is changed as the feature quantity. The time until the actual value follow-up is given. In addition, when the facility diagnosis item is process data actual value monitoring (not directly controlled, only the result obtained), as features, current, temperature, pressure, etc. under a specific sequence condition Can be grouped according to the operating conditions and capture the trend change of the same results that are expected under the same conditions.

  The statistical processing unit 15 performs statistical processing corresponding to the equipment diagnosis item for each group based on the feature amount extracted by the feature amount extraction unit 14. For example, the statistical processing means 15 obtains a statistical amount (average, variance, standard deviation, minimum value, maximum value, threshold value, etc.) for a certain period (level of several days to several weeks) in the latest feature amount. The statistical amount calculated by the statistical processing means 15 is registered in the diagnosis target DB 22.

  The diagnosis unit 16 performs facility diagnosis based on the feature amount extracted by the feature amount extraction unit 14 and the statistical amount processed by the statistical processing unit 15. In the reference value DB 23, a normal feature amount reference value and a statistic reference value corresponding to the equipment diagnosis item are registered for each group. The reference value refers to a past feature value and statistical value obtained for each facility diagnosis item, and is a value representing a past healthy state. The diagnosis means 16 compares the feature amount and the statistic amount corresponding to the equipment diagnosis item for each group with the normal feature amount reference value and the statistic reference value registered in the reference value DB 23, respectively. Check if it has changed more than (allowable).

  The statistic reference value calculation means 17 calculates the statistic reference value using only those having no difference greater than or equal to the predetermined value as a result of the comparison of the statistic by the diagnosis means 16 and updates the reference value DB 23. The statistic reference value calculation means 17 adopts only the feature value of the period determined to have no difference from the past feature value reference value as the feature value reference value, and the statistics value for a certain period (several months to several years level). (Average, variance, standard deviation, minimum value, maximum value, threshold value, etc.) are calculated and used as the statistic reference value.

  The output means 18 outputs the diagnosis result by the diagnosis means 16 by screen display through a display, a speaker, etc., a report to another system, an alarm, guidance in an arbitrary method, and the like. For example, the output unit 18 outputs the target equipment, operation conditions, the degree of difference, etc., for the diagnosis result.

  Next, processing by the equipment diagnosis computer 5 having the above configuration will be described with reference to the flowchart of FIG.

  First, various kinds of master information are registered in the master information DB 20 by the master information registration means 10 (S100), equipment state data is collected by the data collection means 11, and registered in the results DB 21 (S101). Next, the diagnosis item determination means 12 refers to the master information DB 20 and determines the facility diagnosis item from the trigger signal based on the information related to the facility diagnosis item defined in advance and the trigger signal of each facility diagnosis item. The result DB 21 is referred to by the grouping means 13, and the determined equipment diagnosis items are divided into groups based on the group classification defined in advance by the operation conditions (S102). The diagnostic item determination means 12 can also determine a plurality of facility diagnostic items from the same trigger signal by parallel processing.

  Next, the feature quantity extraction unit 14 extracts feature quantities corresponding to the equipment diagnosis items for each group divided by the grouping unit 13, and the feature quantity extraction unit 14 extracts the feature quantities according to the statistical processing unit 15. On the basis of the feature amount, statistical processing corresponding to the equipment diagnosis item is performed for each group divided by the grouping unit 13 (S103). Then, the diagnosis target DB 22 is updated with the feature amount extracted for each group and the calculated statistical amount (S104).

  Next, the diagnosis unit 16 refers to the diagnosis target DB 22 and the reference value DB 23, and the feature amount extracted by the feature amount extraction unit 14 and the statistical amount processed by the statistical processing unit 15 are respectively registered in the reference value DB 23. The facility diagnosis is performed by comparing with a normal feature amount reference value and a statistic reference value and checking whether or not the values have changed by a predetermined value (allowable value) or more (S105).

  Here, if there is no difference greater than the predetermined value, the statistic reference value calculation means 17 calculates the statistic reference value using the latest operation record (S106), and the reference value DB 23 is updated (S107). . On the other hand, if there is a certain difference or more, the output means 18 displays the diagnosis result by the diagnosis means 16 on the screen, display on a speaker or the like, notification to another system, alarm, guidance in an arbitrary method, etc. (S108).

  The above processing (S104 to S107) is performed for each group divided by the grouping unit 13. In addition, the above processing (S104 to S107) needs to be performed for each diagnosis target. However, these processing (S104 to S107) may be performed for each diagnosis target individually. It is also possible to classify by diagnosis target.

  As described above, in the equipment diagnosis computer 5 in the present embodiment, the equipment diagnosis item is determined from the trigger signal based on the information related to the equipment diagnosis item defined in advance and the trigger signal of each equipment diagnosis item, and the determined equipment The diagnosis items are divided into groups based on the group classifications defined in advance by operating conditions, and feature quantities are extracted according to the equipment diagnosis items for each group. The equipment diagnosis is performed by comparing with the feature value reference value in such a case, so the equipment diagnosis can be performed by eliminating the variation of the feature quantity such as the operation time due to the difference in operation conditions, and the diagnosis can be performed with high accuracy. It is.

  In particular, the equipment diagnosis computer 5 performs statistical processing according to equipment diagnosis items for each group based on the extracted feature values, registers the extracted feature values and processed statistics values in the diagnosis target DB 22, The feature value reference value and the statistic reference value in the normal case according to the equipment diagnosis item for each group are registered in the reference value DB 23, and the statistic in accordance with the equipment diagnosis item for each group Because equipment diagnosis is performed by comparing with the reference value, not only the feature quantity, but also the statistical quantity obtained by statistical processing according to the equipment diagnosis item for each group, variation in feature quantity such as operation time due to differences in operating conditions It is possible to perform equipment diagnosis by eliminating the above, and it is possible to perform more accurate diagnosis.

  Further, in the equipment diagnosis computer 5, as a result of the comparison of the statistics, the statistics reference value is calculated using only those having no difference greater than or equal to the predetermined value, and the reference value DB 23 is updated. The reference value is automatically updated, and more reliable equipment diagnosis can be performed.

  The equipment diagnosis apparatus, the equipment diagnosis method, the equipment diagnosis program, and the computer-readable recording medium on which the equipment diagnosis apparatus of the present invention is recorded are useful for diagnosing equipment of a steel plant or other production system that is sequence-controlled. In particular, the present invention is suitable for diagnosing equipment such as an iron and steel plant that has fluctuations in feature quantities such as operation time even in the same sequence due to fluctuations in operating conditions.

1 Equipment 2 PLC
DESCRIPTION OF SYMBOLS 3 General-purpose control network 4 Data collection computer 5 Equipment diagnostic computer 10 Master information registration means 11 Data collection means 12 Diagnostic item determination means 13 Grouping means 14 Feature quantity extraction means 15 Statistical processing means 16 Diagnosis means 17 Statistics reference value calculation means 18 Output means 20 Master information DB
21 Results DB
22 DB for diagnosis
23 Reference value DB

Claims (5)

Diagnostic item determination means for determining a facility diagnosis item from the trigger signal based on information on the trigger diagnosis signal of the facility diagnosis item and each facility diagnosis item defined in advance,
Grouping means for dividing the determined equipment diagnosis items into groups based on group classifications defined in advance by operating conditions;
Feature quantity extraction means for extracting feature quantities according to equipment diagnosis items for each group;
Statistical processing means for performing statistical processing according to equipment diagnosis items for each group based on the feature quantity extracted by the feature quantity extraction means;
A diagnosis target database for registering the feature amount extracted by the feature amount extraction unit and the statistic amount processed by the statistical processing unit;
A reference value database for registering a feature value reference value and a statistic reference value in a normal case according to the equipment diagnosis item for each group,
The feature quantity according to the equipment diagnosis item for each group is compared with the feature quantity reference value in the normal case, and the statistic quantity according to the equipment diagnosis item for each group is further compared with the statistic quantity reference in the normal case. A facility diagnostic apparatus including a diagnostic means for performing facility diagnosis by comparing with a value . The statistic reference value calculation means for calculating the statistic reference value by using only those having no difference of a predetermined value or more as a result of the comparison of the statistic by the diagnostic means, and updating the reference value database. The facility diagnostic apparatus according to 1 . Determining equipment diagnosis items from trigger signals based on information related to predefined equipment diagnosis items and trigger signals of each equipment diagnosis item;
Dividing the determined equipment diagnosis items into groups based on group divisions defined in advance by operating conditions;
Extracting feature quantities according to equipment diagnosis items for each group,
Performing statistical processing according to equipment diagnosis items for each of the groups based on the feature amount;
Registering the feature amount and the statistical amount obtained by the statistical processing in a diagnosis target database;
Registering a feature value reference value and a statistic reference value in a normal case according to the equipment diagnosis item for each group in a reference value database;
Referring to the diagnosis object database and the reference value database, the feature amount corresponding to the facility diagnosis item for each group is compared with a feature amount reference value in a normal case , and further, the facility diagnosis item for each group. A facility diagnosis method including performing facility diagnosis by comparing a statistic corresponding to the statistic with a statistic reference value in a normal case . Diagnostic item determination means for determining a facility diagnosis item from the trigger signal based on information on the trigger diagnosis signal of the facility diagnosis item and each facility diagnosis item defined in advance,
Grouping means for dividing the determined equipment diagnosis items into groups based on group classifications defined in advance by operating conditions;
Feature quantity extraction means for extracting feature quantities according to equipment diagnosis items for each group;
Statistical processing means for performing statistical processing according to equipment diagnosis items for each group based on the feature quantity extracted by the feature quantity extraction means;
A diagnosis target database for registering the feature amount extracted by the feature amount extraction unit and the statistic amount processed by the statistical processing unit;
A reference value database for registering a feature value reference value and a statistic reference value in a normal case according to the equipment diagnosis item for each group,
The feature quantity according to the equipment diagnosis item for each group is compared with the feature quantity reference value in the normal case, and the statistic quantity according to the equipment diagnosis item for each group is further compared with the statistic quantity reference in the normal case. A facility diagnosis program for causing a computer to function as a diagnostic means for performing facility diagnosis by comparing with a value . The computer-readable recording medium which recorded the equipment diagnostic program of Claim 4 .

Images